Anthropic executives flew to Washington, DC, to meet with White House officials Monday in an attempt to resolve a simmering conflict over the company's advanced AI models, according to Wired. After those high-level talks, the two sides remain at odds over the risks posed by Claude Fable 5.
The dispute centers on the Commerce Department's decision to force Anthropic's Mythos — described by Breaking Defense as a bug-hunting AI — along with Fable 5, offline through export controls. The move has drawn sharp criticism from civil liberties groups: the Foundation for Individual Rights and Expression (FIRE) issued a statement condemning the shutdown.
According to Politico, Trump officials met with Anthropic to discuss a potential "truce," but no resolution has been announced. The Hill reports that critics are warning the episode signals dangerously "ad hoc" AI regulation, with no clear legal framework guiding which models can be restricted and why.
Adding an unusual twist, Yahoo Finance and GeekWire report that Amazon CEO Andy Jassy — whose company invested roughly $8 billion in Anthropic — reportedly raised concerns about the Fable models to government officials before the crackdown order was issued.
The fallout is spreading. Breaking Defense notes the Commerce Department's action has complicated the Pentagon's already complicated relationship with Anthropic, a key defense AI supplier. Cybersecurity professionals, meanwhile, are pushing back: according to PYMNTS.com, experts are formally asking federal officials to lift the restrictions on Mythos. Competitor Cohere is reportedly fielding new business inquiries as enterprise customers grow nervous about Anthropic's regulatory exposure, according to StartupHub.ai.
The standoff matters because it is the first high-profile instance of a U.S. administration effectively pulling a frontier AI model off the market, raising urgent questions about who gets to decide which AI is too dangerous — and by what rules.